METHOD AND APPARATUS FOR ALLOCATING RESOURCE REFLECTING ADAPTIVE EVALUATION IN CLOUD COMPUTING FOR HIGH-THROUGHPUT COMPUTING

摘要

Disclosed is a method of efficiently allocating resources in a cloud computing system of the High-Throughput Computing (HTC) field. In more detail, a cloud resource allocation model is proposed in which accumulated job histories are analyzed, a correlation between a performance factor and the job properties is extracted, a factor which is greatly influential in a task performance is identified based on the result of the analyzing and the extracting and is applied to subsequent resource allocations, and reanalysis is performed adaptively.

主权项

1. A method of allocating a cloud resource by a cloud resource management device, the method comprising:
calculating an influence index of each of a factor relating to an application characteristic and a factor relating to a computing resource, by analyzing a previously-performed task profile; selecting a first representative factor having a largest influence index among the factors relating to the application characteristic and a second representative factor having a largest influence index among the factors relating to the computing resource, based on the calculated result; normalizing an application parameter corresponding to the first representative factor among application parameters of a requested task; selecting a task profile having a normalization value equal to the normalized parameter value of the requested task among the previously-performed task profiles, with respect to the first representative factor; selecting one task profile among the selected task profiles based on the second representative factor; allocating a computing resource corresponding to the selected one task profile to the requested task; and after executing the requested task based on the allocated computing resource, evaluating the executed task by comparing an execution time period of the executed task with an execution time period of the selected task profile.